Control Over WirelessHART Network

Control Over WirelessHART Network

S. Han, X. Zhu, Al MokUniversity of Texas at Austin

M. Nixon, T. Blevins, D. ChenEmerson Process Management

Research Scope

2

• WirelessHART Stack Development

• Standard-Compliance Verification

• Localization-aware Applications

• Network Management and Performance Measurement

• Sampling Reduction Techniques and Data Quality Maintenance

• Competition and Collaboration among Protocols in 2.4GHz Band

• Wireless Control

Outline

• Introduction

• Control over WirelessHART network– Control in the Host– Control in the Gateway– Control in the field

• Prototype System and Experiments

• Future Works

3

Introduction• The history of controlling a process plant is

also a history of reducing the number of wires in the industrial plant.

• Applications in process plants– Class C: applications for monitoring– Class B: applications for control– Class A: applications for safety

4

Auxiliary

Critical

The current wireless adoption is at the Class C level.

Introduction (Cont.)

• Control Spectrum in Industrial Plant

• Critical Issues when control goes wireless– Security: open air communication– Reliability: wireless is inherently unreliable– Safety: the topmost concern in a process plant– Speed: Is the speed of WirelessHART enough?– Battery Longevity: replace the battery is costly

5

Supervisory Open Loop Close Loop Critical

Control over WirelessHART Network

6

• Control in the Host– the control module runs in the host’s Function Block

Application

• Control in the Gateway– Gateway needs a function block application layer to allow

configuration and execution of control modules.

• Control in the field– A device supports a set of function blocks.

Control in Host vs. Control in GW

7

• Control in Host– No change to the gateway– Drawback: further communication delay between GW and the Host.

The longer the loop delay is, the worse the control performance.

• Control in the Gateway– Gateway needs to be enhanced with the control modules. – A deterministic schedule is established for all communications by the

network manager.– Function block execution may be fully synchronized with IO

communication.– Gateways may be fully redundant.

Control in GW vs. FF Approach

8

• Control implementation is independent of field device manufacturer. All function blocks are available for use in any control strategy.

• Field devices only need to address measurement/ actuation and communication – minimizing power consumption, simplifying design.

• Control strategy may be fully backed up using redundant gateways.

• The DCS interface to the function block application is simplified since all blocks reside in one device.

• The field devices EDD is very similar to that required for wired HART devices – making it easy for manufacturers to engineer.

What about Control-in-the-Field?

9

• WirelessHART has all of the core features required to support control-in-the-field– Secure peer-to-peer communications– Publish/subscribe capability– Synchronized communications– Fully defined user layer

• WirelessHART is highly optimized and robust– Exception reporting– Mesh behavior

Peer-to-Peer Communications • Devices are allocated a peer-to-peer session

• Network Manager allocates routes and communication resources to achieve reliable and real-time peer-to-peer communications

Measurement in Transmitter Control in Valve

WirelessHART Gateway

Controller

AIOUT

PIDOUT

AOOUT

ININ

IN

BCAL_IN

BCAL_OUTSP

Raw Measurement

Target Position

Actual Position

Publisher Subscriber

Cons of Control-in-the-Field

11

• Force device manufactures to understand control

• Inconsistent implementations of function blocks

• Inconsistent sets of function blocks in devices

• Power Saving Concerns– running function blocks in devices increases battery usage by up

to 3x– running function blocks in controller or GW + leveraging

exception reporting techniques reduces battery usage in devices by up to 20x

WirelessHART Prototype System

12

Major Components in the prototype :

• Network Manager• Gateway• Host Application• Access Point• Field Device• Sniffer

PC Side

Embedded Side

Overall Design of the System

13

Concrete Overview of the System

14

Network Manager Gateway Host Application

Sniffer Access Points

Design of the Gateway

15

Experiment Setup

16

• Network Topology• Gateway + Network Manager• Access Point• Sensor + Actuator

• Control Loop• Sensor publishes the primary value

every 4 seconds• When gateway receives the sensor data,

the Function Block Application issues command 79 to the actuator

Experiment Setup

17

Network Manager Gateway Host Application

Sniffer

SensorActuator

Access Point

Publishing

Control

Experiment Environment

18

• Emerson Process Management Office Building– A lot of non-WirelessHART traffic such as Wi-Fi, Bluetooth.– Nine active WirelessHART networks around

Experimental Results

19

• The test runs for almost 2 hours• Gateway received 1651 burst messages from the sensor• Not a single packet loss is recorded

Observations from Experiments

20

• WirelessHART Supports Control– Experiment is built with actual WirelessHART device, gateway,

and network manager– The environment is noisy– The communication is reliable

• Fastest Loop Achievable?– In theory, we can achieve a 20ms loop period– Two adjacent timeslots are used for publishing and control

respectively

Conclusion

21

• Challenges in introducing wireless into industrial process control

• Comparisons among three approaches for control over WirelessHART networks

• A prototype WirelessHART system for evaluating the performance of control in Gateway approach

Future Work

22

• Measure the jitters in different setups

• Test the fastest loop supported in WirelessHART

• Build a real demo kit